1
Environmental Profile of Algal Hydrothermal Liquefaction – A Country
2
Specific Case Study
3
Bhavish Patel, Miao Guo, Nilay Shah and Klaus Hellgardt*
4
Imperial College London, Department of Chemical Engineering, South Kensington, London, SW7 2AZ, U.K.
5
*Corresponding Author, Tel: +44 (0)20 7594 5577, E-mail: k.hellgardt@imperial.ac.uk
6
Abstract
7
Microalgae are known to be an important feedstock not just for biofuel but also biochemical
8
production. In this investigation we utilise a cradle-to-biorefinery-gate attributional LCA (aLCA)
9
methodology to evaluate the environmental impacts of Nannochloropsis sp. derived algal biocrude
10
production. A database of primary experimental data for continuous fast Hydrothermal Liquefaction
11
(HTL) and Hydrotreating (HDT) is combined with secondary data from literature to investigate the
12
overall environmental profiles of cultivation, dewatering, HTL and HDT for various scenarios based
13
on the energy generation mix of 5 countries (Brazil, UK, Spain, China and Australia) as well as a
14
comparison with fossil crude. The investigation found that Brazil delivers best environmental profiles
15
for all scenarios, primarily due to its significant contribution from hydropower. Furthermore, the
16
cultivation and HTL processes account for nearly 90% of environmental burdens whereas dewatering
17
and HDT only contribute less than 8%. The research findings highlight the importance of the several
18
factors on the resulting 3G biofuel profiling e.g. energy resource, processing technology choice and
19
the co-product(s) and emissions profiling methodology. Algal biocrude is still undergoing research
20
and development compared to the well-developed fossil crude industry. Via integration and
21
optimisation at process and value chain levels, algae-derived biocrude has the potential to deliver an
22
environmentally sustainable alternative to the fossil crude, provided the energy input for processing is
23
from a renewable source.
24
Keywords: Environmental emission profile, HTL, algal biorefinery, LCA
25
Highlights:
26

Cradle to biorefinery-gate LCA using experimental HTL and HDT data
27

Country specific environmental emissions profile for algal biofuel
1
28

Environmental emission comparison of fossil and algal biocrude
29
1.0 Introduction
30
The use of fossil fuels to sustain global energy demand has resulted in detrimental environmental
31
effects. As a major energy consumer, transportation sector alone accounts for 32% of total EU energy
32
demand and in 2012 was responsible for 25% of greenhouse gasses (GHGs) emissions in the EU-28
33
(EU Commission 2013 and EEA 2014). The increasing GHG levels (elevation in CO2 concentration
34
to 400 ppm) along with fossil resources depletion have triggered ambitious policies mandating
35
renewable energy sources within regional/national energy portfolio (Ewald 2013). The role of biofuels
36
in the UK is acknowledged under the EU Renewable Energy Directive (RED), mandating a policy
37
target of a 10% share of renewable energy within the EU transport framework by 2030 (EU Directive
38
2009/28/C). Currently, 1st and 2nd generation biomass feedstock are commercially used for fuel
39
production but due to their intrinsic environmental and social disadvantages (Demirbas 2009),
40
emphasis on 3rd generation (3G) feedstock such as microalgae has resulted in substantial research and
41
development.
42
Compared to other biomass feedstock, microalgae have a faster growth rate, ability to grow in fresh,
43
saline or waste water and more importantly, it is not a primary food crop and thus does not compete
44
for arable land or increase food prices (Patel et al. 2012). Although algae compete with food for
45
fertiliser, the nutrient requirements can be potentially offset by algal cultivation using waste water as
46
well as nutrient recycling from aqueous phase. The lipid, carbohydrate and protein present in
47
microalgae can be converted to fuel precursors or chemicals using several processing technologies
48
(Patel et al. 2015).
49
For conversion of algal biomass without prior fractionation, Hydrothermal Liquefaction (HTL) is
50
understood to be a promising treatment option, particularly since it negates drying thus reducing costs
51
and energy consumption (Lardon et al. 2009). Water under hydrothermal conditions undergoes an
52
enhancement in its properties (density, Cp, ionic product and dielectric constant) causing it to behave
53
like an organic solvent with acidic/basic catalytic properties which is ideal for the conversion of
54
biomass macromolecules to simpler components without addition of harmful solvents (Savage 1999).
2
55
Several studies (Garcia Alba et al. 2011, Jena et al. 2011, Biller and Ross 2013 and Reddy et al. 2011)
56
have successfully demonstrated utilization of HTL to convert algae paste and the research outcome
57
has pointed towards reducing the Residence Time (RT) from hours to minutes (Patel and Hellgardt
58
2013, Jazrawi et al. 2013 and Faeth et al. 2013). Typically, HTL alone is not sufficient and post
59
processing of HTL derived biocrude is necessary to reduce the Oxygen/Nitrogen content through
60
Hydrotreatment (HDT).
61
The implication of combining these processes on environmental performance is expected to be
62
significant especially since there have been major improvements in cultivation and dewatering. It is
63
necessary to examine whether advances in algal biofuel production process is cascaded to result in
64
improvement for its environmental profiling. A widely accepted method used to quantify the
65
environmental metrics of a product/process is Life Cycle Assessment (LCA). There have been
66
several investigations which attempt to quantify LCA impacts of algal biofuels production via
67
different processing routes with most focusing on cultivation or biodiesel production using lipid
68
extraction pathway (Patel et al. 2015). The outcome from these studies elucidated that algae drying
69
process is a main contributor to GHG emission and thus, there has been increasing research attention
70
on wet processing route where algal biomass with high moisture content can be directly utilised as
71
feedstock thus there is no need to dry feedstocks (Chen et al. 2014).
72
To date, a limited number of LCAs have been conducted on HTL (or thermochemical) processing (K.
73
de Boer et al. 2012, Grierson et al. 2013, Fortier et al. 2014, Bennion et al. 2014, Ponnusamy et al.
74
2014, Liu et al. 2013, Frank et al 2013,). Most studies are based on literature data and therefore the
75
general consensus on LCA profiling has not yet been attained. For instance, Frank et al. (2011)
76
concluded that even though HTL uses wet algae, there is an increase of GHGs emission of 1.5 times
77
compared to the traditional route (drying and extraction). But conversely Fortier et al. 2014 suggested
78
that with proper heat integration it is possible to reduce GHGs by 76%. Additionally, by using
79
reaction conditions which maximise biocrude yield (low RT and high temperature) coupled with the
80
high energy recovery in the biocrude of up to 88% (Higher Heating Value (HHV) on a dry weight
81
basis), a further reduction in GHGs and costs can be obtained, resulting in environmental and
82
financial benefits (Zhou et al. 2013). Literature review indicates that no publically available study has
3
83
used data from a continuous HTL flow reactor system which would most probably be implemented at
84
commercial scale.
85
Ultimately, scale-up of an algal biorefinery with multiple product vectors would be the way forward
86
to support (economically) algal biofuel production, but nonetheless, the biofuel produced should instil
87
environmental benefit. Depending on the geographic location or strain used, various products could be
88
extracted and produced using benchmarked processing techniques, for instance HTL biocrude
89
production (Moody et al. 2014, Golberg et al. 2014 and Zinoviev et al. 2015) The algal biorefinery
90
system with energy inputs met by national/regional grid energy mix might not lead to environmentally
91
superior algae biofuel compared with fossil fuels which depends on the region under investigation. A
92
thorough literature review suggests a knowledge gap in LCA of continuous HTL processing (and
93
subsequent HDT) of algal biomass with varying system configuration (e.g. low RT) and different
94
locations (Patel et al. 2015). Particularly there is a lack of country specific comparison analysis of
95
HTL-based algal processing based on empirical data derived from experimental work. This study
96
presents LCA modelling of potential algal biofuel refinery systems to advance the understanding of
97
environmental profiles of algal HTL and HDT processing with varying HTL/HDT reaction
98
configurations and (energy) supply chain at different locations. This study aims to not only fill up the
99
knowledge gap, but more importantly provide scientific insights into the process design and
100
optimisation to enable ongoing empirical work to be more efficiently focused on key environmentally
101
damaging steps to contribute to future sustainable development of algal biorefinery. As such the
102
biorefinery systems are modelled in 5 countries (Brazil, UK, Spain, China and Australia) with
103
differing energy mixture and varying HTL/HDT reaction conditions.
104
2.0 Method
105
2.1 Product system and functional unit
106
A cradle-to-biorefinery-gate attributional LCA (aLCA) approach was applied to evaluate the
107
environmental impacts of the algae-derived liquid fuels (3G biofuel). As illustrated in Figure 1, the
108
subsystems modelled within the system boundary include algae cultivation, algae dewatering, HTL
4
109
and HDT. The aqueous phase fraction produced at the HTL stage is sent to waste water treatment
110
anaerobic digestion (AD) unit. The biogases produced from AD were further sent to combine heat
111
and power (CHP) system for energy recovery.
112
obtained from laboratory experiments (for downstream processing HTL and HDT stages),
113
supplemented by secondary data from literature. However, lab-scale data may not represent the
114
system performance at pilot-scale or commercial-scale fully, given the substantial number of
115
parameters still undetermined. Thus further research would be needed on system scaling-up by
116
process simulation, data production and validation against larger-scale operational data, if available.
117
The LCA study was performed using SimaPro® 7.3 (PhD version) and the biofuel produced from
118
algae, called biocrude, was modelled as potential replacement or blending fuel with crude oil at a
119
refinery and was thus compared directly with fossil crude. The functional unit was defined as ‘per MJ
120
crude oil produced at refinery gate’. A problem oriented (midpoint) approach - CML 2 baseline 2000
121
(v2.05) was applied in the current study as the ‘default’ Life Cycle Impact Assessment (LCIA)
122
method. The impact categories to be investigated include biotic depletion, global warming potential,
123
acidification, eutrophication, ozone depletion, photochemical oxidation and human and eco-toxicities.
124
Other environmental impact categories including land occupation/ land use, Energy Return on
125
Energy Investment (EROEI), are excluded from current LCA system boundary but could be explored
126
in future research. Especially EROEI, as an indicator to assess the energetic profitability of a system,
127
has been applied in algal biocrude research to explore issues like maximizing the energy recoveries
128
of alternatives processes e.g. study carried out by Tercero et al. (2013).EROEI could be incorporated
129
into future LCAs as an energy payback efficiency indicator for algal biocrude production via HTL
130
routes.
131
The acidification characterisation model incorporated in CML 2 baseline 2000 is derived from RAINS
132
model (Regional Air Pollution Information and Simulation) where main acidifying gases accounted
133
for include SOx, NOx, NH3. However, another concern is the ocean acidification effects caused by
134
CO2, which is still in its infancy. Thus further studies are required to explore biological and
LCA inventory was developed using primary data
5
135
biogeochemical consequences of ocean acidification process as well as potential incorporation of
136
ocean acidification evaluation into LCA framework.
Algae Cultivation & Harvesting
Electricity
Algae Dewatering
Electricity
External heat
Solvent
Transportation
CO2
N/P nutrients
Media water
Electricity
Atmospheric Emissions
Water-internal recycle
Algae paste
Atmospheric emissions
Hydrothermal Liquefaction
Waste water
WWT
Biocrude
Bioenergy
Electricity
External heat
H2
Catalyst
Hydrotreating
Atmospheric emissions
Upgraded Biocrude
137
138
Figure 1 - System boundary for algae-derived biocrude
139
2.2 Allocation approach
140
A ‘system expansion’ allocation approach was applied for the biocrude production processes to
141
account for the multiple product mixture present in the system. These were 1) HTL stage where
142
multiple-products include the biocrude oil plus electrical power generated from the on-site AD/CHP
143
system in addition to the nutrient contained in recovered biochar; 2) HDT stage where the upgraded
144
biocrude and biochar recovery as a potential fuel source are produced. It was assumed that the
145
electricity co-product would directly displace an equivalent amount of electrical power generated
146
from the average national grid mixture of the corresponding country in each scenario. The biochar
147
recovery from HTL and HDT were assumed to substitute a ‘functional equivalent’ quantity (dry basis)
6
148
of national average N fertilizer production and an equivalent amount of charcoal, respectively. Thus,
149
this allocation approach awards the biocrude production system with ‘avoided burdens’ credits for the
150
fossil fuel consumption and subsequent emissions avoided for an equivalent amount of avoided
151
product generation.
152
An alternative allocation approach - energy allocation - recommended by the EU RED was examined
153
in sensitivity analysis (Section 3.5) where the environmental burdens were allocated among the co-
154
products (e.g. biocrude and energy recovery) based on their energy contents.
155
A stoichiometric carbon-counting approach was used to ‘track’ the biogenic carbon flows from algae
156
biomass into biocrude oil over the life cycle (Guo et al. 2014). This C-counting approach with regard
157
to the biocrude was applied to firstly determine the carbon ‘sequestered’ into the biocrude (from the
158
algae cultivation phase of the life cycle) and downstream release of this carbon during the subsequent
159
processing of the biocrude. The sequestration of carbon into biocrude thus represents a ‘negative’
160
GHG emission at this stage of the life cycle but this carbon is then returned to the environment in
161
various ways depending upon the consequent fate of the biocrude products.
162
2.3 Process description – Life Cycle Inventory Analysis (LCA)
163
LCA inventory for each process was developed using primary experimental data (for downstream
164
processing HTL and HDT stages), supplemented by literature data. The inventories for chemical
165
production or fuel-specific energy production were derived from the Ecoinvent database (v2.2). The
166
dataset for global fossil crude oil derived from Ecoinvent database (v2.2) was used to represent the
167
average refinery industry for fossil crude oil production including extraction, production and
168
transportation. The detailed LCI analysis is presented below.
169
2.3.1 Cultivation
170
The cultivation data used in this study is derived from literature, whereby it is assumed that the
171
Nannochloropsis sp. was inoculated in a laboratory (tubular reactor) and then cultivated in raceway
172
ponds. Jorquera et al. 2010 investigated the case for algal growth in open raceway ponds with an
173
output of 100000 kg/year of biomass (dry wt. basis) for Nannochloropsis sp. Although algae biomass
7
174
productivity and efficiency could vary with geographical variation in climate condition and resource
175
availability, such special variations has been excluded from LCA scope due to the research focus
176
being placed on algal processing technology evaluation under currently study. A thorough literature
177
review we conducted suggests that limited analyses have been performed at spatially explicit levels
178
(Patel et al. 2015) thus spatial variation in LCA profiles due to the impact of cultivation site location
179
on dominant species and algae growth can be explored in future research. The main energy input
180
associated with the cultivation process is the power for pumping the fluid and paddlewheel for open
181
raceways to give a total energy consumption of 3.785 MJ/kg of dry algal biomass.
182
2.3.2 Dewatering
183
The dewatering of cultivated algal broth is assumed to be conducted through a commercially available
184
spiral plate centrifuge manufactured by Evodus B. V. The use of a centrifuge designed specifically for
185
algae dewatering has the advantage of not damaging the cells in the process and also eliminate the
186
requirement for flocculants, sedimentation or other biomass concentrating procedure. The energy
187
consumption for the equipment is 0.95 kWh per m3 of mixture processed (Evodus 2015). The
188
centrifuge is capable of achieving final concentration of 31.5 wt.-% of Nannochloropsis sp. (Milledge
189
and Heaven 2013) which is within the range of HTL reactor for biomass processing (Knorr et al.
190
2013).
191
2.3.3 HTL
192
The algae paste obtained from the centrifuge was pumped in a HTL reactor at the required pressure
193
and temperature. The HTL reaction data is obtained from experimental investigation using a
194
laboratory scale continuous flow reactor with a volume of 2 ml. Reactions were carried out at 300,
195
325, 350 and 380°C and RT of 0.5, 1, 2 and 4 min using a biomass slurry of 1.5wt.-%. Further
196
detailed information on the experimental investigation and its outcomes can be found in Patel and
197
Hellgardt (2015). The yields for various fractions (biocrude, aqueous soluble fraction and char),
198
energy content (HHV) as well as the elemental (Carbon, Hydrogen, Nitrogen and Oxygen)
199
composition have been summarized in SI 1. It should be noted that although cyclohexane was used in
8
200
the laboratory scale reactor to prevent blockage, solvents are not necessarily required at pilot scale
201
(Jazrawi et al. 2013 and Elliot et al. 2014). Therefore, cyclohexane is not accounted as an input in
202
LCA model. Additionally, the main product after HTL is defined as biocrude, whereas the product
203
from HDT is defined as upgraded biocrude.
204
Based on experimental data, multiple regression analyses were performed using built-in Microsoft
205
Excel regression analysis toolbox to generate regression models for LCA inputs in Simapro.
206
Continuous variables (e.g. yields of biocrude, aqueous fraction and char, elemental compositions)
207
were expressed as regression functions of HTL residence time and reaction temperature – this
208
approach would allow for development of a dynamic LCA model with projection power.
209
Experimental data and sample multivariable regression model can be found in SI 1. It is assumed that
210
the aqueous soluble fraction produced at the HTL stage is fed into AD and CHP system for energy
211
recovery. The aqueous soluble fraction is defined as the matter partitioned into the aqueous phase
212
after HTL. In this study, all biodegradable compounds were assumed to be digested under AD thus
213
theoretical biogas production was used to estimate the maximum energy recovery. The biogas
214
production (assumed as 65% CH4 (v/v); 35% CO2 (v/v)) from AD system was estimated based on the
215
theoretical chemical oxygen demand (COD), which can be derived from Equation 1, and the
216
theoretical CH4 potential (COD equivalence of CH4), which was calculated based on Equation 2
217
(Speece 1996). The energy generation from CHP system is calculated based on a study carried out on
218
an AD commercial plant in the UK, where electricity conversion efficiency is approximately 1.2
219
kWh/m3 biogas (net calorific value of biogas is 21.48 kJ/L), and 50% of energy contained in biogas is
220
assumed to be recovered as thermal energy (Guo et al. 2013) . For the formula CnHaObNc (1mol)
221
COD =
222
Where, the formula of the aqueous soluble fraction was derived from elemental analysis results and n,
223
a, c and b refer to the moles of C, H, O and N, respectively.
2  n  (a  3  c) / 2  b
 32 g O2
2
(1)
224
225
9
226
CH4 + 2O2  CO2 + 2H2O
227
Where - each mole of CH4 consumes two moles of oxygen. Therefore, 1g COD destruction is
228
equivalent to 0.35L CH4 at 0°C and 760mm Hg pressure (STP) or 0.395L CH4 at 35°C and one
229
atmosphere (Speece 1996).
230
Although the algal concentration in the reaction solution for the experimental investigation was low, it
231
is possible for a HTL reactor system to handle 36.6 wt.-% solids (Knorr et al. 2013) without any
232
adverse effect from increased algal concentration to HTL (Barreiro et al. 2013 and Toor et al. 2011).
233
For the energy input required for HTL we used the Q=mCpT relationship for algal paste with biomass
234
concentration of 31.5 wt.-% to give a temperature dependent value ranging from 1.87 to 2.42 MJ/kg
235
biomass processed. The ambient temperature was taken as 25°C and the Cp of algae as 1.371 kJ/kg/K
236
(Ponnusamy et al. 2014).
237
2.3.4 HDT
238
Catalytic HDT is the last biorefinery process under investigation. The biocrude product from HTL is
239
subjected to catalytic upgrading under 110 bar of hydrogen at 400°C and RT of 1hr to yield upgraded
240
biocrude which could potentially be blended with fossil crude at a refinery. The aim of the HDT
241
process was to increase the HHV of the oil by deoxygenation. Further information on the
242
experimental procedure can be found in Patel et al. 2014 and the data used is included in SI 1. The
243
results of 5wt.-% loaded Pt/Al2O3 (surface area 101.36 m2/g and pore volume 0.2194 cm3/g ) Pt/C
244
(surface area 1204.39 m2/g and pore volume 0.8742 cm3/g) and control (uncatalysed) were used as
245
model experiments since these gave the highest deoxygenation, HHV and upgraded biocrude yield,
246
respectively [SI 1] based on our laboratory experiments. The energy requirement for HDT was
247
estimated assuming crude oil equivalent heat capacity of biocrude as 1.81 kJ/kg/K (Burger et al. 1985)
248
and a ΔT of 340°C based on HTL outlet temperature (after potential heat recovery and fraction
249
separation) of 60°C. In addition to this, the H2 requirement was assumed as 0.0375 wt.-% of biocrude
250
(Schmidt et al. 2014).
251
2.4 LCA Case Description
(2)
10
252
Several cases accounting for different experimental conditions, processing options and allocation
253
approach were explored for the five countries. In Case 1, cumulative LCIA profiles of HTL processes
254
under all reaction conditions were explored followed by contributional analyses of three reaction
255
conditions (300°C/0.5 min, 380°C/0.5 min and 380°C/4 min), which were selected to represent HTL
256
reaction severity (upper and lower bound) and best-performance process with highest biocrude yield.
257
The energy content, defined as the Higher Heating Value (HHV) was not of particular interest during
258
the selection since most HTL reaction conditions yield similar HHV values. Hence, according to our
259
previous work (Patel and Hellgardt 2015) HTL at 380°C and 0.5 min RT is considered as favourable
260
reaction condition thus selected as illustrative example in cases 3-5 due to its high biocrude yield and
261
good aqueous soluble fraction recovery. In Case 2, the biocrude derived from selected HTL reaction
262
conditions was compared to fossil crude on an equivalent function unit basis. Case 3 was
263
implemented to portray the process contributional analysis for upgraded biocrude oil produced from
264
HTL reacted at 380°C and 0.5 min RT as well as selected HDT reaction conditions. In Case 4, a
265
comparison between upgraded biocrude produced at three HDT conditions and fossil crude is
266
presented. Finally in Case 5 a scenario sensitivity analysis method was applied which involves
267
calculating different scenarios in order to analyse the influences of input parameters on either LCIA
268
output results or rankings (Guo et al. 2014). A reversal of the rank order of counterparts for LCA
269
comparisons and an arbitrary level of a 10% change in the characterized LCIA profiles for a single
270
product system were chosen as the sensitivity threshold, above which the influence of allocation
271
approach or AD scenario was considered to be significant.
272
3.0 Results and Discussion
273
The following sections present and discuss the results of the various scenarios modelled in this
274
investigation along with the impact categories and normalised comparisons (%) presented in Figures
275
2-9. The LCIA scores for each individual impact category and scenarios are given in SI 2 (Tables S1-
276
S52).
277
3.1 Case 1- Contributional analysis for biocrude oil at HTL stage
11
278
Cumulative environmental profiles for the biocrude is presented in Figure 2 where the overall
279
contribution of electrical energy inputs at algae cultivation and HTL stages vary within the range of
280
55-65% and 32-40%, respectively. Energy demand for dewatering is responsible for up to 7% of
281
environmental impact across all impact categories whereas less than 3% of environmental burden is
282
attributed to N/P nutrient required for algae cultivation. These can be explained by the fossil fuel
283
consumptions as well as the emissions evolved from fossil resource extraction and transportation,
284
electricity generation and transmission. On abiotic depletion, natural gas, hard coal and crude oil
285
inputs dominate the environmental profiles. Scores on GWP100, acidification and POCP are driven
286
by the emissions from combustion of coal, natural gas and heavy fuel oil including SOx, NOx, CO,
287
CO2, and CH4. Organic compounds like PAH and heavy metals such as Arsenic, Nickel, Selenium
288
released from fuel combustion or production of power plant construction materials account for over
289
85% of the impact on human toxicity. Over 90% of aquatic toxic impact can be attributed to the heavy
290
metallic elements/ions released to water body from ash/spoil disposal during coal mining and
291
combustion (e.g. nickel ion, vanadium ion, copper ion, cobalt, and beryllium). Not only electricity
292
generation but also its transmission brings environmental damage on terrestrial toxicity impact
293
category due to the chromium emitted at transmission network and also the atmospheric emission of
294
mercury evolved from coal combustion. Fossil resource production and transport share more than
295
95% of the ODP load, which is mainly caused by CClBrF2 emission from natural gas transportation,
296
CClF3 released from crude oil production and CCl4 as a result of Chlorine inputs for coal production.
297
The environmental benefits arising from biogenic carbon sequestration at algae cultivation stage is
298
sufficient to offset the positive impacts induced by biocrude production (HTL with a residence time of
299
0.5 min, except in Australia) leading to biocrude products with negative GHG scores (Figure 2).
300
Exported electricity from the HTL process and nutrient recovery from biochar also add towards
301
‘avoided burden’ credits across all impact categories by substitution for the equivalent amounts of
302
electricity generated from the respective national grids and ‘functional equivalent’ quantity (dry basis)
303
of national average N fertilizer production. As shown in Figure 2, HTL at a residence time of 0.5 min
304
(300/380 °C) had higher nutrient recovery (biochar yield) and greater export of surplus electricity due
12
305
to higher amount of matter in the aqueous phase resulting in higher COD content in liquid being sent
306
to WWT compared to HTL at 380°C, 4 min. However, these benefits were overridden by
307
environmental burdens in most cases, except in the case of biocrude production in Brazil, UK, Spain
308
and China in GWP100 terms (Figure 2 and 3).
309
The comparisons between five countries are determined by the fossil resources for electricity
310
production mix. Grid electricity in Brazil is predominantly generated by hydropower which accounts
311
for 80%, whereas Spain and UK have a relatively balanced fuel mix, but conversely Australia and
312
China are coal intensive energy generators. Therefore, Brazil followed by Spain and UK are
313
suggested as better locations for producing HTL biocrude based on the environmentally advantageous
314
figures compared to China and Australia on almost all impact categories (Figure 2) except for ODP,
315
where the comparison results are driven by profile of natural gas and fuel oil.
316
As expected, the reaction temperature and RT play an important role in LCA outcomes. For short
317
residence time (0.5 min or 2 min) an improvement in environmental performance with increased HTL
318
temperature is observed on almost all impact categories except for GWP100 where no clear trend is
319
shown, but the optimal reaction temperature identified for GHG balance is 300°C. With longer
320
residence time (4 min), different trends are suggested in Figure 3 which elucidates a decline in
321
environmental scores with increased HTL temperature. For HTL with reaction temperature above
322
300°C, a RT of 0.5 min is indicated as an environmentally superior option compared with longer RT.
323
Similarly, for HTL at lower temperature (300°C), 4 min shows several environmental advantages over
324
shorter RTs. An exception is Brazil where a RT of 0.5 min demonstrates clear advantage for any
325
reaction temperature group. HTL at 300°C for 0.5 min deliver the best GWP100 profiles, whereas on
326
other impact categories HTL at 380°C for 0.5 min represents the optimum option offering the best
327
environmental performance (Figure 3), which can be explained by its high biomass conversion
328
efficiency (biocrude yield ) .
329
330
13
331
332
14
333
Figure 2 - LCIA profiles of biocrude at HTL stage at 300°C, 4 min (a), 380°C, 0.5 min (b) and
334
380°C, 4 min (c). [Unit: 1kg biocrude; Method: CML 2 baseline 2000]
335
336
Figure 3 - LCIA profiles of biocrude at HTL stage in Brazil (a), Australia (b), China (c), Spain
337
(d) and UK (e). [Unit: 1kg biocrude at HDT stage; Method: CML 2 baseline 2000]
338
3.2 Case 2 - Cumulative life cycle impacts for HTL biocrude vs. fossil crude
339
The LCA comparison between biocrude produced via HTL and fossil crude is performed on an
340
equivalent function unit basis (Figure 4). Under most HTL reaction conditions (except for 380 °C, 4
341
min), and countries (except for Australia) investigated, HTL processed biocrude is observed to be
15
342
overall environmentally superior to fossil crude on GWP100 and ODP impact category basis.
343
However, higher impact scores than fossil crude oil are found in the other impact categories (except
344
for Brazil, which benefits from its hydropower, delivers better scores on abiotic depletion, and
345
eutrophication). In this study, the dataset derived from Ecoinvent (V2.2) representing global average
346
refinery industry for fossil crude production (including extraction, production and transportation) has
347
been adopted in the LCA comparisons. However, country-specific environmental profiles for fossil
348
crud accounting for spatial variation in transportation distance is not accounted for and could be
349
explored in future LCA research as it is beyond the scope of this investigation.
350
16
351
Figure 4 - LCIA comparison of HTL biocrude vs. fossil crude at 300°C, 4 min (a), 380°C, 0.5
352
min (b) and 380°C, 4 min (c). [Unit: 1 MJ crude oil produced; Method: CML 2 baseline 2000]
353
3.3 Case 3 - Contributional analysis for upgraded biocrude oil at HDT stage
354
The ‘cradle-to-biorefinery gate’ LCIA profiles of upgraded biocrude using different catalyst for HDT
355
is presented in Figure 5, where biocrude derived from HTL with a RT of 0.5 min and a reaction
356
temperature of 380°C is used as an illustrative example. Regardless of different countries and reaction
357
conditions, LCIA profiles of upgraded biocrude is dominated by energy-intensive cultivation and
358
HTL processes. These are responsible for over 90% of environmental burdens across all impact
359
categories mainly due to the fossil fuel consumed for electricity generation and the emissions released
360
during fossil fuel production, transportation and combustion. These environmental burdens include
361
the fossil resource depletion caused by coal, natural gas and crude oil inputs for grid electricity
362
generation, impacts on GWP100, acidification and POCP due to atmospheric emission of CH4, CO2
363
SOx NOx released from fossil fuel combustion. Additionally, the eutrophication precursors such as
364
PO43- , NOx from coal extraction and combustion, ODP impacts brought by CClBrF2 and CClF3
365
emissions from natural gas transportation and crude oil extraction, toxic compounds and heavy metals
366
released from electricity generation and transmission (e.g. Beryllium, Vanadium, Nickel, PAH
367
evolved from coal mining and burning and heavy fuel oil combustion as well as the Chromium
368
emissions at electricity transmission stage) all contribute towards substantiating the environmental
369
burden. Less than 8% of overall environmental burden is attributable to energy demand at dewatering
370
and HDT process. Generally, the contribution of H2 inputs at HDT stage and nutrient demand for
371
algae cultivation (diammonium phosphate and urea) are negligible, except for abiotic depletion where
372
H2 component is responsible for up to 20% of environmental burden (positive value above line) due to
373
the gas and oil consumed for H2 production i.e. cracking process. The burden from H2 production
374
could potentially be reduced by using renewable H2 from solar/wind water splitting or biomass
375
gasification with a lower environmental emission profile (Koroneos et al. 2004). Biogenic carbon
376
sequestered into upgraded biocrude at the algae cultivation stage brings significant ‘negative’ impacts
377
on GWP100, acting to ‘offset’ the impact of burden incurred from the biocrude production and
17
378
leading to Brazilian case with a net negative GHG balance at biorefinery factory gate. Environmental
379
‘savings’ (Figure 5) across all impact categories are also derived from the ‘avoided burden’ credits
380
from electricity exported from AD/CHP system during the HTL process and biochar recovery at HTL
381
and HDT stages (substitutions for the functional equivalent amount of fertilizer nutrients and charcoal
382
respectively).
383
Amongst the five countries, Brazil and Spain represent good locations for producing upgraded
384
biocrude, delivering lowest environmental impact on almost all impact fronts (Figure 5) except for
385
ODP, where China leads. These outcomes are driven by the different fossil resources for electricity
386
production mix. Brazil and Spain benefit from hydropower driven electricity supply and a balanced
387
fuel resource, respectively. Conversely, Australia and China where grid electricity supplies are
388
dominated by coal represent the worst scenarios in most impact categories except for low ODP
389
impacts of China, which can be explained by minor proportion of natural gas and fuel oil in electricity
390
production mix.
391
The reaction conditions as well as conversion efficiency/yield is another key factor affecting
392
environmental profiles for biocrude production system. Benefiting from high upgraded biocrude yield,
393
non-catalytic HDT (control) delivers a better environmental profile than catalytic HDT (with Pt/Al2O3
394
or Pt/C) (Figure 6). The production process with HTL reaction condition of 380°C and 0.5 min,
395
combined with a non-catalytic HDT deliver the best environmental performances compared with
396
other systems for producing upgraded biocrude (Figure 6). The lack of environmentally advantageous
397
outcome for catalytic HDT process suggests the need of better catalyst capable of producing upgraded
398
biocrude with better HHV at higher yields and greater denitrogenation as well as optimising the
399
reaction conditions.
400
401
402
18
403
19
404
Figure 5 - ‘Cradle-to-biorefinery-gate’ LCIA profiles of biocrude at HDT stage for control (a),
405
Pt/Al2O3 (b), and Pt/C (c) catalysed reactions. [Unit: 1kg upgraded biocrude; Method: CML 2
406
baseline 2000]
407
408
Figure 6 - ‘Cradle-to-biorefinery-gate’ LCIA profiles of upgraded biocrude in Spain (a), UK (b),
409
China (c), Brazil (d) and Australia (e). [Unit: 1kg biocrude at HDT stage; Method: CML 2
410
baseline 2000)
411
3.4 Case 4 - Cumulative life cycle impacts for upgraded biocrude oil at use phase
20
412
As illustrated in Figure 7, comparison of results (based on energy functional units) between upgraded
413
biocrude and fossil crude vary with countries, reaction conditions, and impact categories. Upgraded
414
biocrude produced in Brazil is shown as environmentally competitive to fossil crude in terms of
415
GWP100, abiotic depletion, ODP and POCP, especially for HTL condition of 380°C/0.5 min and
416
HDT (non-catalytic) which shows environment advantage or equivalent over fossil crude for nearly
417
all impact categories. With the reduction in conversion efficiency (biomass to biocrude) during HTL
418
(defined as either increased RT to 4 min or decreased temperature to 300°C), upgraded biocrude
419
derived via catalytic HDT in Spain and UK shift to a disadvantageous ODP position with regard to
420
fossil crude whereas GWP100 profiles of biocrude produced in Spain move from being negative (net
421
GHG removal from atmosphere) to positive values (net addition of GHG to atmosphere). In general,
422
to deliver the same energy functional unit, biocrude struggles and is hardly competitive with fossil
423
fuel on acidification, eutrophication and toxicity impact categories.
21
424
425
Figure 7 - LCIA comparison of upgraded biocrude vs. fossil crude at 300°C, 0.5 min (a), 380°C,
426
0.5 min (b) and 380°C, 4 min (c). [Unit: 1 MJ crude oil produced; Method: CML 2 baseline
427
2000)
428
22
429
3.5 Case 5 - Sensitivity analysis on allocation approach and AD scenarios
430
Sensitivity analysis using the allocation approach (Figure 8) indicated that the influence of allocation
431
choice on LCIA profiles of biocrude produced from HTL varies with reaction conditions modelled for
432
the impact categories investigated. GWP100 was the most sensitive impact category for the allocation
433
approach but it is not the case for remaining impact categories. Switching from system expansion to
434
the energy allocation approach leads to significantly increased GWP100 score (range from 20% to
435
over 100%) for biocrude modelled for most HTL reaction conditions whereas GWP100 profiles
436
remain stable in the case of severe HTL reaction at 380°C/4 min. With energy allocation approach,
437
China moves from being negative (net GHG removal from atmosphere) to positive value (net addition
438
of GHG to atmosphere), which is above the chosen 10% sensitivity threshold. The allocation
439
approach was not a sensitive factor in terms of the LCIA comparison between HTL biocrude and
440
fossil crude in almost all impact categories but in GWP100 where China shifts from environmentally
441
advantageous to disadvantageous over fossil crude.
442
Furthermore, sensitivity analysis was conducted to test the influences of AD scenario at HTL stage on
443
the LCIA comparisons of HTL biocrude vs. fossil crude. As demonstrated in Figure 9 (and Tables
444
S50-S52 in SI2 (in comparison with results in Figure 4 and Table S21-23 in SI2), the LCIA profiles of
445
HTL biocrude in general is sensitive to AD scenarios. Shifting from HTL with AD energy recovery to
446
exclusion of AD unit, the increase in the environmental impacts for HTL biocurde vary with HTL
447
reaction conditions, country and impact categories (generally ranging between 5-55%). With
448
increasing reaction severity of HTL, the effects of AD energy recovery on the LCIA profiles of
449
biocrude decreases significantly. AD scenario does not produce any effect on the environmental
450
footprint of HTL for the reaction at 380°C and RT 4 min. GWP100 represents the most sensitive
451
impact category, where the case of China moves from negative to positive for the GWP100 scores
452
when the AD energy recovery unit is excluded. AD case study is only a sensitive parameter for the
453
comparisons between fossil crude and the biocrude supply chain modelled for China (China case
454
study shifted from superior to inferior system to fossil crude) but not for other LCIA comparisons.
23
455
456
Figure 8 - Sensitivity analysis of characterized LCIA profiles of fossil crude vs. HTL biocrude at
457
HTL 300°C, 0.5 min (a), 380°C, 0.5 min (b) and 380°C, 4 min (c) (unit: 1 MJ crude oil
458
produced; method: CML 2 baseline 2000).
24
459
460
25
461
Figure 9 - Sensitivity analysis on AD scenarios – fossil crude vs. HTL biocrude at 300°C, 0.5 min
462
(a), 380°C, 0.5 min (b) and 380°C, 4 min (c) (unit: 1 MJ crude oil produced; method: CML 2
463
baseline 2000).
464
4.0 Conclusion
465
The utilisation of algae as biomass for biofuel production can only be realised at an industrial
466
biorefinery scale if it is economically and environmentally sound. Economic feasibility can be
467
achieved by value chemical production but environmental improvement measured using assessment
468
tools such as LCA can only be achieved by improvement in process integration and technological
469
development. As determined from this investigation, even with significant reduction in HTL
470
processing time, HTL still accounts for over 40% environmental burdens in almost all impact
471
categories. From our investigation it can be seen that if the energy mix used to power an algal
472
biorefinery is environmentally beneficial, substantial improvement can be made in reducing GWP100
473
emissions, as seen for the case of Brazil where up to 80% electricity is generated using hydropower.
474
Furthermore, the results indicate the need to transition towards cleaner electricity production systems
475
to ensure the benefit is cascaded down to emerging liquid fuel production technologies such as algae.
476
If implemented, whether the higher costs (and environmental benefits) associated with fuel produced
477
from algae mask the lower costs (and environmental detriments) associated with fossil crude remains
478
to be seen. Specifically, the key aspect of the comparative analyses presented here for algae derived
479
biocrude oil production across various potential supply chains has been to highlight the importance of
480
the following main factors on the resulting biofuel profiles-.
481

The specific energy resource being used (e.g. hydropower dominated electricity in Brazil)
482

HTL processing technology (e.g. residence time, reaction temperature)
483

Level of biocrude upgrading (e.g. severity of upgrading at HDT stage)
484

Importance of co-product(s) and emissions profiling methodology applied in the LCA
485
methodology (e.g. system expansion vs. energy allocation approach)
26
486
Under most reaction conditions biocrude produced (except for Australia) via HTL delivers
487
environmental saving compared with fossil crude on GWP100 and ODP; whereas at HDT stage, only
488
upgraded biocrude produced in Brazil remained environmentally competitive to fossil resource in
489
terms of GWP100 and ODP profiles. Algae feedstock processed at the optimal configurations
490
(e.g.380°C, 0.5 min) with environmentally favourable energy supply (e.g. Brazil scenarios) is shown
491
in our modelling to offer life cycle GWP100 savings over fossil crude of 80% or more, placing them
492
well within the most desirable categories being targeted by policymakers internationally (e.g. the EU
493
Renewable Energy Directive, the USA Renewable Fuel Standard). However, looking at the overall
494
GWP100 (Table 1), it becomes apparent that although a substantial contributor, the negative impact of
495
HTL can be marginalized by energy recovery of by-products, which was AD in this case. Whether
496
this can be realised effectively at an industrial scale remains to be seen and can only be explored
497
through further practical investigation.
498
Table 1 – GWP100 for HTL, AD and Total overall impact [Total is based on process defined in
499
Fig. 1. Only HTL and AD shown for comparison purpose]
500
501
502
503
Country
Brazil
UK
Spain
China
Australia
HTL
AD
Total
GWP 100 (kg CO2 eq per kg biocrude)
0.1020
-0.3690
-4.8800
1.4
-0.7320
-1.59
0.993
-0.6080
-2.6
1.98
-1.0800
-0.132
2.12
-0.9110
0.349
504
505
Algal biocrude is still under development and research compared to the well-developed fossil crude
506
industry. Modelling research as demonstrated in current study could provide analytical tools and
507
insightful information for process or value chain configuration and enables ongoing empirical work to
508
be more efficiently focused on key performance-limiting and environmentally damaging steps to
509
accelerate algal biorefinery research. The LCA of HDT/HTL reaction matrix with varying conditions
510
investigated under this study highlight optimal configuration for HTL/HDT process. By combining
27
511
statistical method (e.g. multivariate regression analysis) with LCA model, this study also
512
demonstrates the development of a dynamic model with projection power. More research effort is
513
required to explore various optimisation options for algal biorefinery, including feedstock
514
optimisation (e.g. strain screening), process integration and optimisation (e.g. energy integration,
515
energy and resource recovery), as well as supply chain integration and optimisation (e.g.
516
incorporation of algal biocrude system into other production supply chains, selection of optimal
517
location for biorefinery). In the long term, such integration and optimisation at process and value
518
chain level, combined with the utility of renewable energy source for processing, it is likely that
519
algae-derived biocrude has the potential to exhibit environmentally sustainable biorefinery.
520
521
522
523
524
525
526
527
528
529
530
531
532
28
533
Abbreviations
3G - Third Generation
ABD - Abiotic Depletion
ACD - Acidification
EUT - Eutrophification
FAE - Freshwater Aquatic Ecotoxicity
GWP - Global Warming Potential
HDT - Hydrotreatment
HTL - Hydrothermal Liquefaction
HUT - Human Toxicity
LCA - Life Cycle Assessment
MAE - Marine Aquatic Ecotoxicity
ODP - Ozone Layer Depletion
POCP - Photochemical Oxidation Potential
RT - Residence Time
TEE - Terrestial Ecotoxicity
534
535
536
537
538
539
540
541
542
543
544
545
29
546
547
548
References
1. Azadi, P., Brownbridge, G., Mosbach, S., Smallbone, A., Bhave, A., Inderwildi, O., & Kraft,
549
M. (2014). The carbon footprint and non-renewable energy demand of algae-derived
550
biodiesel. Applied Energy, 113, 1632-1644.
551
552
2. Barreiro, DL., Prins,W., Ronsse, F., & Brilman, W., (2013). Hydrothermal liquefaction
553
(HTL) of microalgae for biofuel production: State of the art review and future prospects,
554
Biomass and Bioenergy. 53, Pages 113-127
555
556
3. Batan, L., Quinn, J., Willson, B., & Bradley, T. (2010). Net energy and greenhouse gas
557
emission evaluation of biodiesel derived from microalgae. Environmental science &
558
technology, 44(20), 7975-7980.
559
560
4. Bennion, E. P., Ginosar, D. M., Moses, J., Agblevor, F., & Quinn, J. C. (2015). Lifecycle
561
assessment of microalgae to biofuel: Comparison of thermochemical processing pathways.
562
Applied Energy.
563
564
565
5. Biller, P., & Ross, A. B., (2012). Hydrothermal processing of algal biomass for the
production of biofuels and chemicals. Biofuels. 3, 603-623.
566
567
568
6. Bradley, T., Maga, D., & Antón, S. (2015). Unified approach to Life Cycle Assessment
between three unique algae biofuel facilities. Applied Energy.
569
570
571
7. Burger, J., Sourieau, P.,& Combarnous, M., 1985. Thermal methods of oil recovery, Technip
, Paris, 1985
572
30
573
574
8. Campbell, P. K., Beer, T., & Batten, D. (2011). Life cycle assessment of biodiesel production
from microalgae in ponds. Bioresource technology, 102(1), 50-56.
575
576
577
9. Chen, W. H., Lin, B. J., Huang, M. Y., & Chang, J. S. (2014). Thermochemical conversion of
microalgal biomass into biofuels: A review. Bioresource technology, 184, 314-327.
578
579
10. Davis, R. E., Fishman, D. B., Frank, E. D., Johnson, M. C., Jones, S. B., Kinchin, C. M.,
580
Skaggs, R. L., Venteris, E. R., & Wigmosta, M. S. (2014). Integrated evaluation of cost,
581
emissions, and resource potential for algal biofuels at the national scale. Environmental
582
science & technology, 48(10), 6035-6042.
583
584
11. de Boer, K., Moheimani, N. R., Borowitzka, M. A., & Bahri, P. A. (2012). Extraction and
585
conversion pathways for microalgae to biodiesel: a review focused on energy consumption.
586
Journal of Applied Phycology, 24(6), 1681-1698.
587
588
589
12. Demirbas, A., 2009. Pyrolysis of biomass for fuels and chemicals. Energy Sources Part A. 31,
1028-1037.
590
591
13. EEA, EEA greenhouse gas - data viewer. 2014, The European Environment Agency. URL:
592
http://www.eea.europa.eu/data-and-maps/data/data-viewers/greenhouse-gases-viewer
593
[Accessed Aug 2014]
594
595
14. European commission Consumption of energy. URL:
596
http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Consumption_of_energy
597
[Accessed Aug 2014]
598
31
599
600
601
602
15. European Union, Directive 2009/28/EC of the European Parliament and of the Council 2009,
The European Parliament and the Council of the European Union
16. Evodus Website , http://www.evodos.eu/index.php/spiral-plate-technology.html [Accessed
02/02/2015]
603
604
605
17. Ewald, J., 2013. Carbon Dioxide at NOAA’s Mauna Loa Observatory reaches new milestone:
Tops 400 ppm’. NOAA Research, 10 May 2013.
606
607
608
18. Faeth, J. L., Valdez, P. J., Savage, P. E., 2013. Fast hydrothermal liquefaction of
Nannochloropsis sp. to produce biocrude. Energy Fuels. 27, 1391-1398.
609
610
19. Fortier, M. O. P., Roberts, G. W., Stagg-Williams, S. M., & Sturm, B. S. (2014). Life cycle
611
assessment of bio-jet fuel from hydrothermal liquefaction of microalgae. Applied Energy,
612
122, 73-82.
613
614
20. Frank, E. D., Elgowainy, A., Han, J., & Wang, Z. (2013). Life cycle comparison of
615
hydrothermal liquefaction and lipid extraction pathways to renewable diesel from algae.
616
Mitigation and Adaptation Strategies for Global Change, 18(1), 137-158.
617
618
21. Garcia Alba, L., Torri, C., Samorì, C., van der Spek, J., Fabbri, D., Kersten, S. R., Brilman,
619
D. W., 2011. Hydrothermal treatment (HTT) of microalgae: evaluation of the process as
620
conversion method in an algae biorefinery concept. Energy 458 Fuels. 26, 642-657.
621
622
22. Grierson, S., Strezov, V., & Bengtsson, J. (2013). Life cycle assessment of a microalgae
623
biomass cultivation, bio-oil extraction and pyrolysis processing regime. Algal Research, 2(3),
624
299-311.
625
32
626
23. Guo, M., Littlewood, J., Joyce, J., Murphy, R., 2014. The environmental profile of bioethanol
627
produced from current and potential future poplar feedstocks in the EU. Green Chemistry. 16,
628
4680-4695.
629
630
24. Guo, M., Stuckey, DC., Murphy, RJ., 2013. Is it possible to develop biopolymer production
631
systems independent of fossil fuels? Case study in energy profiling of polyhydroxybutyrate-
632
valerate (PHBV). Green Chem. 15, 706-717.
633
634
635
25. Jazrawi, C., Biller, P., Ross, A. B., Montoya, A., Maschmeyer, T., Haynes, B. S., 2013. Pilot
plant testing of continuous hydrothermal liquefaction of microalgae. Algal Res. 2, 268-277.
636
637
26. Jena, U., Das, K. C., Kastner, J. R., 2011. Effect of operating conditions of thermochemical
638
liquefaction on biocrude production from Spirulina platensis. Bioresour. Technol. 102, 6221-
639
6229
640
641
642
27. Jonker, J. G. G., & Faaij, A. P. C. (2013). Techno-economic assessment of micro-algae as
feedstock for renewable bio-energy production. Applied Energy, 102, 461-475.
643
644
28. Jorquera, O., Kiperstock, A., Sales, E. A., Embiruçu, M., Ghirardi, M. L., 2010. Comparative
645
energy life-cycle analyses of microalgal biomass production in open ponds and
646
photobioreactors. Bioresour. Technol. 101, 1406-1413
647
648
29. Knorr, D., Lukas, J., Schoen, P. 2013. Production of Advanced Biofuels via Liquefaction –
649
Hydrothermal Liquefaction Reactor Design: April 5, 2013. NREL/SR-5100-60462, National
650
Renewable Energy Laboratory, Golden, Colorado, USA.
651
33
652
30. Koroneos, C., Dompros, A., Roumbas, G., & Moussiopoulos, N. (2004). Life cycle
653
assessment of hydrogen fuel production processes. International Journal of Hydrogen Energy,
654
29(14), 1443-1450
655
656
657
31. Lardon, L., Hélias, A., Sialve, B., Steyer, J. P., Bernard, O., 2009. Life-cycle assessment of
biodiesel production from microalgae. Environ. Sci. Technol., 43, 6475-6481.
658
659
32. Liu, X., Saydah, B., Eranki, P., Colosi, L. M., Mitchell, B. G., Rhodes, J., & Clarens, A. F.
660
(2013). Pilot-scale data provide enhanced estimates of the life cycle energy and emissions
661
profile of algae biofuels produced via hydrothermal liquefaction. Bioresource technology,
662
148, 163-171.
663
664
665
33. Malik, A., Lenzen, M., Ralph, P. J., & Tamburic, B. (2014). Hybrid life-cycle assessment of
algal biofuel production. Bioresource Technology, 184, 436-443.
666
667
668
34. Milledge, JJ., Heaven, S., 2013. A review of the harvesting of micro-algae for biofuel
production. Reviews in Environ.Sci. Biotech. 12, 165-78.
669
670
35. Mu, D., Min, M., Krohn, B., Mullins, K. A., Ruan, R., & Hill, J. (2014). Life Cycle
671
Environmental Impacts of Wastewater-Based Algal Biofuels. Environmental science &
672
technology, 48(19), 11696-11704.
673
674
36. Orfield, N. D., Fang, A. J., Valdez, P. J., Nelson, M. C., Savage, P. E., Lin, X. N., &
675
Keoleian, G. A. (2014). Life cycle design of an algal biorefinery featuring hydrothermal
676
liquefaction: effect of reaction conditions and an alternative pathway including microbial
677
regrowth. ACS Sustainable Chemistry & Engineering, 2(4), 867-874.
678
34
679
37. Passell, H., Dhaliwal, H., Reno, M., Wu, B., Amotz, A. B., Ivry, E., Gay, M., Czartoski, T.,
680
Laurin, L., & Ayer, N. (2013). Algae biodiesel life cycle assessment using current
681
commercial data. Journal of environmental management, 129, 103-111.
682
683
684
685
38. Pate, R., Klise, G., & Wu, B. (2011). Resource demand implications for US algae biofuels
production scale-up. Applied Energy, 88(10), 3377-3388.
686
687
39. Patel, B., Arrillaga, P. A., Izadpanah, A., Millan-Agorio, M., & Hellgardt, K., (2015)
688
Catalytic Hydrotreatment of fast liquefaction algal biocrude. Energy & Fuels (awaiting
689
submission)
690
691
692
40. Patel, B., Hellgardt, K., 2013. Hydrothermal upgrading of algae paste: Application of
31
P‐NMR. Environ. Prog.Sustain.Energy. 32, 1002-1012.
693
694
695
41. Patel, B., Tamburic, B., Zemichael, F. W., Dechatiwongse, P., Hellgardt, K., 2012. Algal
biofuels: a credible prospective?. ISRN Renew. Energy 2012
696
697
698
42. Patel. B., Hellgardt, K., 2015. Hydrothermal Upgrading of Algae Paste in a Continuous Flow
Reactor. Bioresour. Technol. 19, 460-468
699
700
43. Patel. B., Guo. M., Izadpanah. A., Shah. N., Hellgardt. K., 2015. A Review on Hydrothermal
701
Pre-treatment Technologies and Environmental Profiles of Algal Biomass Processing.
702
Bioresour. Technol. (In Print)
703
35
704
44. Ponnusamy, H., Reddy, H, K., Muppaneni, T., Downes, C, M., Deng, S., 2014. Life cycle
705
assessment of biodiesel production from algal bio-crude oils extracted under subcritical water
706
conditions. Bioresour. Technol.170, 450-461.
707
708
45. Quinn, J. C., & Davis, R. (2014). The potentials and challenges of algae based biofuels: A
709
review of the techno-economic, life cycle, and resource assessment modeling. Bioresource
710
technology.
711
712
46. Razon, L. F., & Tan, R. R. (2011). Net energy analysis of the production of biodiesel and
713
biogas from the microalgae: Haematococcus pluvialis and Nannochloropsis. Applied Energy,
714
88(10), 3507-3514.
715
716
47. Reddy, H. K., Muppaneni, T., Rastegary, J., Shirazi, S. A., Ghassemi, A., & Deng, S., 2013.
717
Hydrothermal extraction and characterization of bio‐crude oils from wet chlorella
718
sorokiniana and dunaliella tertiolecta. Environ. Prog.Sustain.Energy, 32, 910-915.
719
720
721
48. Sander, K., & Murthy, G. S. (2010). Life cycle analysis of algae biodiesel. The International
Journal of Life Cycle Assessment, 15(7), 704-714.
722
723
724
49. Savage, P. E., 1999. Organic chemical reactions in supercritical water. Chme. Rev. 99, 603622.
725
726
50. Schmidt, A. J., Albrecht, K. O., Hart, T. R., Butcher, M. G., Drennan, C., Snowden-Swan, L.
727
J., Kinchin, C. (2014). Process design and economics for the conversion of algal biomass to
728
hydrocarbons: whole algae hydrothermal liquefaction and upgrading. Pacific Northwest
729
National Laboratory.
730
36
731
732
51. Singh, A., & Olsen, S. I. (2011). A critical review of biochemical conversion, sustainability
and life cycle assessment of algal biofuels. Applied Energy, 88(10), 3548-3555.
733
734
735
52. Speece, R, E., 1996. Anaerobic Biotechnology for Industrial Wastewaters, Archae Press,
Nashville, Tennessee, USA.
736
737
53. Stephenson, A. L., Kazamia, E., Dennis, J. S., Howe, C. J., Scott, S. A., & Smith, A. G.
738
(2010). Life-cycle assessment of potential algal biodiesel production in the United Kingdom:
739
a comparison of raceways and air-lift tubular bioreactors. Energy & Fuels, 24(7), 4062-4077.
740
741
54. Toor, SS., Reddy, H., Deng, S., Hoffmann, J., Spangsmark, D., Madsen, LB., Holm-
742
Nielsen, JB., Rosendahl, LA., 2013. Hydrothermal liquefaction of Spirulina and
743
Nannochloropsis salina under subcritical and supercritical water conditions, Bioresour.
744
Techno. 131, 413-419
745
746
55. Tredici, M. R., Bassi, N., Prussi, M., Biondi, N., Rodolfi, L., Zittelli, G. C., & Sampietro, G.
747
(2015). Energy balance of algal biomass production in a 1-ha “Green Wall Panel” plant: How
748
to produce algal biomass in a closed reactor achieving a high Net Energy Ratio. Applied
749
Energy.
750
751
56. Zhou, Y., Schideman, L., Yu, G., & Zhang, Y. (2013). A synergistic combination of algal
752
wastewater treatment and hydrothermal biofuel production maximized by nutrient and carbon
753
recycling. Energy & Environmental Science, 6(12), 3765-3779.
754
37